September 9, 2013

Mimicry, the link between crime and cocooning trends

Criminals can be thought of as predators, and their victims as prey. Elsewhere (mostly in comments) I've described the basic dynamic between these two groups that generates cycles in both crime rates and trust levels. To recap:

First, people (i.e., potential victims, not criminals) become more trusting of others. Higher trust levels allow criminals to gain access to the prey, who are basically defenseless once access has been gained. So, second, crime rates begin rising a bit after trust levels do. Rising crime rates make people more wary of others. Hence, third, after crime rates have been going up for awhile, average people begin to withdraw their trust of others. Falling trust levels make it more difficult for criminals to gain access to victims, so fourth and finally, crime rates begin to plummet a bit after trust levels begin to fall.

Of course, after crime rates have been falling for so long, average people sense less of a reason to keep others at arm's length and under suspicion. So trust levels begin rising for the first time in a long while. Now we're back to where we started, and the crime-and-trust cycle repeats itself.

I've never been able to find quite the right analogy, but it finally hit me that the theory and observations about mimicry in biology are the most relevant. Criminals don't wear conspicuous signals or stand out in obvious ways to the average person. They do just the opposite -- appear to be basically trustworthy, but will take advantage of others for personal gain when they feel like it, and not feel any compunction.

Sociopaths are "aggressive mimics" of normal, trustworthy people (the "model"), and they exploit those who let their guard down around them (the "dupes"). Some of this is visual -- dressing and grooming themselves a normal way, speaking in the tone of voice of a charming person interested in other people, using the body language of a normal person during the approach toward and around the prey (until the pounce).

They don't have to mimic the most trustworthy of all people -- they just have to convince the dupes that they're basically trustworthy, so that they won't raise their guard around them. Maybe it's someone who doesn't look particularly charismatic and dashing, but hey, why should that keep you from opening your door when he knocks at your door pleading desperately (and convincingly) for help? OK, so he doesn't look like he's going to be canonized as a saint some day, but hey, why should that keep you from letting him babysit your child or serve as his Boy Scout troop leader?

Most people who seem trustworthy are trustworthy. Maybe that should be said the other way around: trustworthy people tend to have certain identifying traits (mostly honest signals), and we come to judge others by the presence or absence of those traits. So in general, there's little harm in letting our guard down around someone who feels trustworthy.

But some signals are capable of being faked, and anyone who can fake them will find success in exploiting average people. As their prevalence increases, though, normal people become more aware of them, causing the predators to grow at a slower and slower rate, as trust levels begin to plateau instead of continuing up up up. This is a case of frequency-dependent selection.

When people withdraw into their cocoons, it's not only a physical isolation -- even when out and interacting with others, they have their guard up. So the analogy from biology is not where the prey merely avoid spaces where would-be predators could snag them, or honing their escape/evasive maneuvers  (although that's going on too).

Cocooning folks show some fear of physical places that could be hot spots of crime ("this place looks sketchy"). But their predominant fear seems to be an anxiety that somebody they trust, or give a pass to, will turn out to be a sociopath who will take advantage of them. It's fear of deception, not of being attacked by someone you didn't even know was there. That means they're more worried about (aggressive) mimics than about stealthy and physically imposing predators.

Importantly, people are not worried about dangerous members of some foreign group swooping in to prey on them. That's more like a war zone. Crime is more local and caused by folks who look like your neighbors. Don't they always say that, about the guy who shot his wife and children one day after work -- neighbors tell reporters that he seemed like a normal, even model husband and father? No one could have told? If you can't even trust your own blood, who can you trust?

That's what worries people, not ghetto blacks riding over the hill. For another thing, most people don't live near ghetto blacks or Mexicans, either in the U.S. or in all the other Western countries that saw a crime wave from roughly 1960 to 1990, or from 1580 to 1630, or from 1780 to 1830. Crime rates rose across all 50 states, not just those with large black populations. The changing mood and response to rising crime also affected all 50 states.

To pick one example from the list in the Wikipedia article, the closest parallel looks like the system consisting of a grouper fish (the dupe) that allows a cleaner fish (the model) to get near it in order to eat parasites off the grouper's fins, while a species that mimics the cleaner fish is allowed close but then takes a bite out of the grouper's fin, enjoying a nice little meal for itself, and harming rather than helping the grouper.

The grouper and cleaner fish are in a mutualistic relationship, and that's how normal people within the same society interact with each other. The sociopath is one who can closely imitate pro-social behavior (body language, speech, dress and grooming, etc.), and take a quick bite out of a trusting person before splitting and moving on to target another unsuspecting dupe.

When the grouper is preyed on in this way by the mimic species, it raises its guard. That precludes some of the beneficial interactions it could have had from letting the true cleaner fish get close. But letting go of those benefits is just the price you have to pay these days to avoid those damn mimics. Normal people are also aware of how deprived their lives have become during the cocooning period of the past 20 or so years (or back during the mid-century). But they figure it's the price you have to pay to avoid being used by a deceiver.

Only when people feel such a wide mismatch between their anxiety about deception and their own real-world experience (and second-hand learning), will they figure the price is no longer worth it. Like, when was the last time I or anybody I know was victimized by a sociopath? If it's been that remote, maybe I can let my guard down more around strangers -- after all, there don't seem to be that many mimics out there these days.

I think the human case has something to tell the broader biological theory of mimicry, or at least how it's presented to the public. Namely, that mimicry is unstable and will rise and fall in cycles. Typically when you read about mimicry, it's portrayed as a static thing (though obviously allowing for new cases to be discovered). The hoverflies that mimic the appearance of nasty wasps, and thereby ward off predators, must have been doing so since forever. Those cuckoos that lay eggs mimicking the eggs of the host bird species, must have been doing that since forever. That's just what they do.

Well, no it isn't. When we look for examples of mimicry now, we're only going to find those cases that are like a rising-crime period in human history. In all likelihood, 1000 years ago, 10,000 years ago, or however long ago, there were mimicry arrangements that no longer exist today, even though the mimic, model, and dupe species are still with us. It's just that, in the intervening time, the dupe wised up and mimicry no longer paid off. Presumably the same will happen with the species targeted by cuckoo birds.

The only exception is the common side-blotched lizards, where every review of their dynamics focuses on the cyclical nature of mimicry. But that ought to come up in every example presented -- when and how does the dupe catch on, and what happens to the mimic when the jig is up? Do they evolve to mimic another dupe species, close enough to the first to require minimal change? Or do they just dwindle in number? In whatever records we have of the history of these species (like fossils), do we see a period where the mimic had not set on the dupe just yet? If so, when did it all begin? Also in the record, do we see clear signs of mimicry that are no longer present in the descendant species?

These cyclical dynamics are way more fascinating than the snapshot of a successful mimicry case, where we just have a good laugh at the dupe's expense. Same with humans -- how societies cycle over time is way more fascinating than how people interact with each other during a snapshot.


  1. Is there literature on cycles of mimicry in other species? I had also been under the impression that cuckoo mimicry was fairly stable. But since those are two different species, frequency dependent selection may not apply the same way.

    You've written about tattoos before, Steve has data from Australia on what kind of person gets them.

  2. I should have mentioned "red queen theory" since my default mental model for predator/parasite-prey co-evolution.

  3. Giving it further thought, under frequency dependent selection where the fitness of each allele is a negatively sloped linear function of its frequency, there should be theoretical frequencies at which the fitness of each allele is equal. That's an equilibrium and we'd have stasis rather than cycles. That's the standard sort of explanation I'd read of frequency dependent selection. In order to get cycles we'd need something like inertia that keeps a shift going even as force builds against it.

  4. My boss is a sociopath. I cannot look him the eye most of the time i talk to him, because he constantly reads the emotions on my face and tries to copy them. The funny part is, when i laugh or smile, he doesnt seem able to experience a genuine enough happiness or empathy to share the emotion. A smile seems to translate to a half sneer.

    P.S., agnostic, did you see the new thor movie?

    The imagery could have used some blur, foreground, background, etc. the constant sharpness is so boring. If i paid you a dollar for every majestic flyover instead of showing the city from a 6' human perspective...

  5. "did you see the new thor movie?"

    I considered seeing The Avengers, but decided against it. The Thor movie wasn't even a consideration.

  6. "under frequency dependent selection where the fitness of each allele is a negatively sloped linear function of its frequency"

    In that case, fitness is maximized at frequency = 0, and it would never move away from 0.

    Treating mimicry as a form of predator-prey or host-parasite interactions captures what's going on better. The mimic knocks into a dupe and gains in number, while the dupe declines. Without mimics, dupes grow in a standard way (exponential, logistic, whatever). Without dupes to exploit, a world full of mimics declines monotonically (exponential decay, or whatever).

    That gives you the boom-and-bust dynamics that generate cycles. The Lotka-Volterra equations are the simplest model of the general approach.

  7. The other good thing about ecological models is that you don't have to worry about genetics, heritability, plasticity, etc. The label "mimic" tells you what their role in the system is, regardless of what made them that way.


You MUST enter a nickname with the "Name/URL" option if you're not signed in. We can't follow who is saying what if everyone is "Anonymous."